antigravity
Pass
Audited by Gen Agent Trust Hub on Apr 4, 2026
Risk Level: SAFECOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill facilitates the execution of the Antigravity CLI to delegate tasks or obtain second opinions. This capability involves invoking an external binary to process data based on agent inferences.
- [PROMPT_INJECTION]: The skill represents an indirect prompt injection surface because it uses untrusted user prompts to infer execution modes and construct commands. Ingestion points: The skill analyzes the user prompt context at runtime to determine delegation vs validation modes. Boundary markers: No delimiters or ignore-embedded-instructions warnings are defined in the SKILL.md to protect the CLI execution logic. Capability inventory: The agent is instructed to use the Antigravity CLI and its associated flags based on inferred context. Sanitization: The instructions do not specify any validation or sanitization steps for the user-provided data before it is passed to the CLI tool.
Audit Metadata